Better haproxy 503 page #12409
Better haproxy 503 page #12409
Conversation
|
@sspeiche please take a look at this proposal, if it looks good we'll get it merged to be in 3.5. |
|
The thing I struggle with a bit is, thinking about an end user of OpenShift (a developer deploying apps): "What do I do as the owner of the application that is getting this message?". I believe you get this message if you have service port that doesn't match the pods. Wouldn't you get a 404 if host doesn't exist? Guess not sure what host is in OpenShift terms and how a user would overcome this problem. I'm a good example of someone who can deploy an application but when given this 503 message, I'd be a bit lost on what to do to fix it. |
|
You only get the 503 error from the haproxy when you hit our router but for
an invalid host (or path).
I suppose suggesting that they check the hostname and path as the first
item might not be bad for the end-users. The other cases are for when the
developer tries to hit their app and it fails.
…-ben
On Mon, Jan 9, 2017 at 2:37 PM, Steve Speicher ***@***.***> wrote:
The thing I struggle with a bit is, thinking about an end user of
OpenShift (a developer deploying apps): "What do I do as the owner of the
application that is getting this message?".
Then I think if it as an application end user, what do I do with this
information.
I believe you get this message if you have service port that doesn't match
the pods.
Wouldn't you get a 404 if host doesn't exist? Guess not sure what host is
in OpenShift terms and how a user would overcome this problem.
I'm a good example of someone who can deploy an application but when given
this 503 message, I'd be a bit lost on what to do to fix it.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#12409 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AJkavXaVulUbduRJDr2BjFeoHAtSRvyrks5rQox9gaJpZM4Lc8MD>
.
|
|
Potentially both app developer and end-users of deployed apps can hit this page, so I think we must provide some troubleshooting info for developers but at the same time can't be too much technical. What about something like this (feel free to fine-tune the wording): Possible reasons you're seeing this page:
|
|
I think is good. Do we have any doc links we'd want to include in case a user says "I don't know what this is saying?" or is that getting crazy? |
|
We can't include doc links because we can't brand this page, its not
customizable so in online / dedicated it would be pointing to the wrong
docs.
…On Tue, Jan 17, 2017 at 12:49 PM, Steve Speicher ***@***.***> wrote:
I think is good. Do we have any doc links we'd want to include in case a
user says "I don't know what this is saying?" or is that getting crazy?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#12409 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/ABZk7W8AtmB5e4cf3NXZrsqbYOBBMee6ks5rTP88gaJpZM4Lc8MD>
.
|
|
@sspeiche unfortunately we can't add doc links here because this page is not customizable for the Origin/OCP/etc differences. |
|
Updated, @sg00dwin please review? |
fabianofranz
force-pushed
the
827_error_pages
branch
from
d93e4ba to
b0c863d
Compare
There was a problem hiding this comment.
@fabianofranz There's one extraneous rule to be removed, otherwise looks good to me.
| } | ||
| a { | ||
| color: ; |
There was a problem hiding this comment.
This a {...} rule can be removed.
fabianofranz
force-pushed
the
827_error_pages
branch
from
b0c863d to
a7faf10
Compare
|
@sg00dwin fixed, tks! [merge] |
|
continuous-integration/openshift-jenkins/merge SUCCESS (https://ci.openshift.redhat.com/jenkins/job/test_pr_origin/12966/) (Image: devenv-rhel7_5703) |
|
Evaluated for origin merge up to a7faf10 |
|
[Test]ing while waiting on the merge queue |
|
Evaluated for origin test up to a7faf10 |
|
continuous-integration/openshift-jenkins/test SUCCESS (https://ci.openshift.redhat.com/jenkins/job/test_pr_origin/12966/) (Base Commit: eaa36ed) |
Improves styling and information in the haproxy router error page.